Solution(Assignment 3)
Nikhil Revankar - 01FB16ECS230
Vikram G - 01FB16ECS484
Vinayaka Kamath - 01FB16ECS445
Assignment Number - 3
Team Name: Turing Machines
Question 1
The data is cleaned for invalid months and preprocessed to feed it into the plotly functions.
library(plotly)
library(zoo)
data<-read.csv("~/Documents/Assignments/DA/week3_Assign/RainfallData.csv",header=TRUE)
data<-subset(data,substr(data$YYYYMM,5,6)!="13")
states = setNames(unique(subset(data,select = c("Description"))),c("State"))
rownames(states)=c()
dataset = unique(subset(data,select = c("YYYYMM")))
for(i in 1:nrow(states))
{
indata=setNames(subset(data,select=c("YYYYMM","Value"),Description==states[c(i),]),c("YYYYMM",as.character(states[c(i),])))
rownames(indata)=c()
dataset = merge(dataset,indata,by="YYYYMM")
}
dataset <- data.frame(lapply(dataset, function(x) {gsub("Not Available",0,x)}))
colnames(dataset)[3] <- "Tamil_Nadu"
colnames(dataset)[7] <- "Madhya_Pradesh"
colnames(dataset)[9] <- "West_Bengal"
colnames(dataset)[10] <- "Arunachal_Pradesh"
plot_ly(dataset, x = ~as.Date(as.yearmon(as.character(dataset$YYYYMM), "%Y%m"), frac = 1)) %>%
add_lines(y = ~as.vector(dataset$Karnataka), name = "Karnataka") %>%
add_lines(y = ~as.vector(dataset$Tamil_Nadu), name = "Tamil Nadu", visible = F) %>%
add_lines(y = ~as.vector(dataset$Kerala), name = "Kerala", visible = F) %>%
add_lines(y = ~as.vector(dataset$Orissa), name = "Orissa", visible = F) %>%
add_lines(y = ~as.vector(dataset$Maharashtra), name = "Maharashtra", visible = F) %>%
add_lines(y = ~as.vector(dataset$Madhya_Pradesh), name = "Madhya Pradesh", visible = F) %>%
add_lines(y = ~as.vector(dataset$Bihar), name = "Bihar", visible = F) %>%
add_lines(y = ~as.vector(dataset$West_Bengal), name = "West Bengal", visible = F) %>%
add_lines(y = ~as.numeric(dataset$Arunachal_Pradesh), name = "Arunachal pradesh", visible = F) %>%
layout(
title = "Rainfall in India",
xaxis = list(rangeslider = list(type = "date"),title="year"),
yaxis = list(title = "Rainfall"),
updatemenus = list(
list(
y = 0.3,
buttons = list(
list(method = "restyle",
args = list("visible", list(TRUE, FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE)),
label = "Karnataka"),
list(method = "restyle",
args = list("visible", list(FALSE, TRUE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE)),
label = "Tamil Naidu"),
list(method = "restyle",
args = list("visible", list(FALSE, FALSE, TRUE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE)),
label = "Kerala"),
list(method = "restyle",
args = list("visible", list(FALSE, FALSE, FALSE,TRUE,FALSE,FALSE,FALSE,FALSE,FALSE)),
label = "Orissa"),
list(method = "restyle",
args = list("visible", list(FALSE, FALSE, FALSE,FALSE,TRUE,FALSE,FALSE,FALSE,FALSE)),
label = "Maharashtra"),
list(method = "restyle",
args = list("visible", list(FALSE, FALSE, FALSE,FALSE,FALSE,TRUE,FALSE,FALSE,FALSE)),
label = "Madya Pradesh"),
list(method = "restyle",
args = list("visible", list(FALSE, FALSE, FALSE,FALSE,FALSE,FALSE,TRUE,FALSE,FALSE)),
label = "Bihar"),
list(method = "restyle",
args = list("visible", list(FALSE, FALSE, FALSE,FALSE,FALSE,FALSE,FALSE,TRUE,FALSE)),
label = "West Bengal"),
list(method = "restyle",
args = list("visible", list(FALSE, FALSE, FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,TRUE)),
label = "Arunachal Pradesh")
))
)
)Question 2
A grouped bar graph as well as a line graph was plotted to understand the distribution of top 5 crops with respect to the extent of land on which they were harvested. The crop “Total foodgrain” is dropped due to a lot of N/A records being associated with respect to this entity. The grouped bar graphs is plotted to isolate each year and observe values for different crops in that year. Double click on the legend to isolate a year.
data<-read.csv(file="~/Documents/Assignments/DA/week3_Assign/TNAgri.csv", header=TRUE, sep=",")
data <- subset(data,Crop!="Total foodgrain")
crops_data = subset(data,select = c('Crop','Crop_Year','Area'))
Years_crops = aggregate(Area~Crop+Crop_Year,crops_data,sum)
top5_data = aggregate(Area~Crop,Years_crops,sum)
sorted_data = top5_data[with(top5_data, order(-Area)), ][c(1:5),]
interested_data <- data.frame(subset(Years_crops,Crop %in% sorted_data$Crop))
#bar plot
data1997 = subset(interested_data,Crop_Year == 1997)
data1998 = subset(interested_data,Crop_Year == 1998)
data1999 = subset(interested_data,Crop_Year == 1999)
data2000 = subset(interested_data,Crop_Year == 2000)
data2001 = subset(interested_data,Crop_Year == 2001)
data2002 = subset(interested_data,Crop_Year == 2002)
data2003 = subset(interested_data,Crop_Year == 2003)
data2004 = subset(interested_data,Crop_Year == 2004)
data2005 = subset(interested_data,Crop_Year == 2005)
data2006 = subset(interested_data,Crop_Year == 2006)
data2007 = subset(interested_data,Crop_Year == 2007)
data2008 = subset(interested_data,Crop_Year == 2008)
data2009 = subset(interested_data,Crop_Year == 2009)
data2010 = subset(interested_data,Crop_Year == 2010)
data2011 = subset(interested_data,Crop_Year == 2011)
data2012 = subset(interested_data,Crop_Year == 2012)
data2013 = subset(interested_data,Crop_Year == 2013)
barplt <-
plot_ly() %>%
add_trace(x = factor(data1997$Crop), y = data1997$Area,name=1997, type = 'bar',
text =data1997$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(164, 219, 194)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data1998$Crop), y = data1998$Area,name=1998, type = 'bar',
text =data1998$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(49, 249, 18)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data1999$Crop), y = data1999$Area,name=1999, type = 'bar',
text =data1999$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(249, 18, 249)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data2000$Crop), y = data2000$Area,name=2000, type = 'bar',
text =data2000$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(30, 252, 248)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data2001$Crop), y = data2001$Area,name=2001, type = 'bar',
text =data2001$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(183, 162, 0)',
line = list(color = 'rgb(8,48,107)', width =1.5))) %>%
add_trace(x = factor(data2002$Crop), y = data2002$Area,name=2002, type = 'bar',
text =data2002$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(252, 63, 30)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data2003$Crop), y = data2003$Area,name=2003, type = 'bar',
text =data2003$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(10, 153, 65)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data2004$Crop), y = data2004$Area,name=2004, type = 'bar',
text =data2004$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(198, 14, 239)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data2005$Crop), y = data2005$Area,name=2005, type = 'bar',
text =data2005$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(255, 122, 45)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data2006$Crop), y = data2006$Area,name=2006, type = 'bar',
text =data2006$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(223, 255, 45)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data2007$Crop), y = data2007$Area,name=2007, type = 'bar',
text =data2007$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(45, 146, 255)',
line = list(color = 'rgb(19,38,17)', width = 1.5))) %>%
add_trace(x = factor(data2008$Crop), y = data2008$Area,name=2008, type = 'bar',
text =data2008$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(231, 196, 255)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data2009$Crop), y = data2009$Area,name=2009, type = 'bar',
text =data2009$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(155, 48, 107)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data2010$Crop), y = data2010$Area,name=2010, type = 'bar',
text =data2010$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(130, 190, 255)',
line = list(color = 'rgb(88,248,107)', width = 1.5))) %>%
add_trace(x = factor(data2011$Crop), y = data2011$Area,name=2011, type = 'bar',
text =data2011$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(160, 141, 99)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data2012$Crop), y = data2012$Area,name=2012, type = 'bar',
text =data2012$Crop_Year , textposition = 'auto',
marker = list(color = 'rgb(88, 66, 229)',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
add_trace(x = factor(data2013$Crop), y = data2013$Area,name=2013, type = 'bar',
text =data2013$Crop_Year , textposition = 'auto',
marker = list(color = '#AED581',
line = list(color = 'rgb(8,48,107)', width = 1.5))) %>%
layout(title = "Analysis of trends for top 5 Crops over the years",
barmode = 'group',
yaxis = list(title = c("Area")))
barplt#line plot
Groundnut <- subset(interested_data,Crop == 'Groundnut')
Jowar <- subset(interested_data,Crop == 'Jowar')
Rice <- subset(interested_data,Crop == 'Rice')
Sugarcane <- subset(interested_data,Crop == 'Sugarcane')
Urad<- subset(interested_data,Crop == 'Urad')
Years = unique(subset(interested_data,select = c('Crop_Year')))
lineplot <- plot_ly(type = 'scatter', mode = 'lines') %>%
add_trace(x=~Groundnut$Crop_Year, y =~Groundnut$Area, name = 'Groundnut', line = list(color = 'rgb(249, 151, 22)', width = 4)) %>%
add_trace(x=~Jowar$Crop_Year, y =~Jowar$Area, name = 'Jowar', line = list(color = 'rgb(128, 2, 255)', width = 4)) %>%
add_trace(x=~Rice$Crop_Year, y =~Rice$Area, name = 'Rice', line = list(color = 'rgb(15, 255, 15)', width = 4)) %>%
add_trace(x=~Sugarcane$Crop_Year, y =~Sugarcane$Area, name = 'Sugarcane', line = list(color = 'rgb(255, 15, 123)', width = 4)) %>%
add_trace(x=~Urad$Crop_Year, y =~Urad$Area, name = 'Urad', line = list(color = 'rgb(255, 255, 2)', width = 4)) %>%
layout(title = "Analysis of trends for top 5 Crops over the years",
xaxis = list(title = "Years"),
yaxis = list (title = "Area"))
lineplotQuestion 3
library(data.table)
graph<-read.csv("~/Documents/Assignments/DA/week3_Assign/GunViolence.csv",header=TRUE)
states_code<-data.frame(state.name, state.abb,stringsAsFactors = FALSE)
DT<-data.table(subset(graph,select=c(state,n_killed,n_injured)))
g<-data.table(subset(graph,select=c(incident_id,state)))
a=data.frame(g[, .(number_of_incidents = uniqueN(incident_id)), by=state])
b=data.frame(DT[, lapply(.SD,sum), by=state])
preprocess=merge(a,b)
df=merge(preprocess,states_code,by.x="state",by.y="state.name")
df$hover <- with(df, paste(state, "<br>",
"Number of kills", n_killed,
"<br>", "Number of injuries",n_injured))
l <- list(color = toRGB("white"), width = 2)
g <- list(
scope = 'usa',
projection = list(type = 'albers usa'),
showlakes = TRUE,
lakecolor = toRGB('white')
)
plot_geo(df, locationmode = 'USA-states') %>%
add_trace(
z = ~number_of_incidents, text = ~hover, locations = ~state.abb,
color = ~number_of_incidents, colors = 'Reds'
) %>%
colorbar(title = "Count") %>%
layout(
title = 'Crimes in USA by State<br>(Hover for breakdown)',
geo = g
)Question 4
Two word clouds and one pie chart each for a gender is drawn. The first word cloud represents the resolution topics while the second one is created from the tweet data. They are preprocessed using stemming and the stop words are removed. The pie charts help to represent the share of each category in the whole data.
setwd('~/Documents/Assignments/DA/week3_Assign/')
data<-read.csv(file="NewYearResolution15.csv", header=TRUE, sep=",")
male_data = subset(data,select = c("name","gender","Resolution_Category"),gender=="male")
female_data = subset(data,select = c("name","gender","Resolution_Category"),gender=="female")
female_count = setNames(aggregate(gender~Resolution_Category,female_data,length),c("Category","Count"))
male_count = setNames(aggregate(gender~Resolution_Category,male_data,length),c("Category","Count"))
#pie charts
colors <- c('rgb(229, 90, 66)','rgb(125, 229, 66)','rgb(66, 229, 220)','rgb(229, 66, 223)','rgb(93, 66, 229)','rgb(255, 251, 20)','rgb(20, 255, 184)','rgb(184, 119, 214)','rgb(202, 214, 119)','rgb(168, 127, 97)')
female_plot <-plot_ly(female_count,labels = ~Category, values = ~Count,
textposition = 'inside',
textinfo = 'percent',
insidetextfont = list(color = '#00000'),
marker = list(colors = colors,
line = list(color = '#FFFFFF', width = 1))) %>%
add_pie(hole = 0.3) %>%
layout(title = "Resolution Category Proportion (Female)", showlegend = T,
xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))
female_plotmale_plot <-plot_ly(male_count,labels = ~Category, values = ~Count,
textposition = 'inside',
textinfo = 'percent',
insidetextfont = list(color = '#00000'),
marker = list(colors = colors,
line = list(color = '#FFFFFF', width = 1))) %>%
add_pie(hole = 0.3) %>%
layout(title = "Resolution Category Proportion (Male)", showlegend = T,
xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))
male_plot#install.packages("tm") # for text mining
#install.packages("SnowballC") # for text stemming
#install.packages("wordcloud") # word-cloud generator
#install.packages("RColorBrewer") # color palettes
library(tm)
library(SnowballC)
library(RColorBrewer)
library(wordcloud)
#resolution_topics wordcloud
topic.Corpus<-Corpus(VectorSource(data$resolution_topics))
topic.Clean<-tm_map(topic.Corpus, PlainTextDocument)
topic.Clean<-tm_map(topic.Corpus,tolower)
topic.Clean<-tm_map(topic.Clean,removeNumbers)
topic.Clean<-tm_map(topic.Clean,removeWords,stopwords("english"))
topic.Clean<-tm_map(topic.Clean,removePunctuation)
topic.Clean<-tm_map(topic.Clean,stripWhitespace)
dtm <- TermDocumentMatrix(topic.Clean)
rtm <- as.matrix(dtm)
rtv <- sort(rowSums(rtm),decreasing=TRUE)
rtd <- data.frame(word = names(rtv),freq=rtv)
wordcloud(words = rtd$word,freq = rtd$freq,max.words =200 ,random.color = TRUE,random.order=FALSE,rot.per=0.55,
colors=brewer.pal(8, "Dark2"))
#tweet text wordcloud
#text.Corpus<-Corpus(VectorSource(data$text))
text.Corpus<-Corpus(VectorSource(gsub("#NewYearsResolution", "\\1", data$text)))
text.Clean<-tm_map(text.Corpus, PlainTextDocument)
text.Clean<-tm_map(text.Corpus,tolower)
text.Clean<-tm_map(text.Clean,removeNumbers)
text.Clean<-tm_map(text.Clean,removeWords,stopwords("english"))
text.Clean<-tm_map(text.Clean,removePunctuation)
text.Clean<-tm_map(text.Clean,stripWhitespace)
dtm <- TermDocumentMatrix(text.Clean)
tm <- as.matrix(dtm)
tv <- sort(rowSums(tm),decreasing=TRUE)
td <- data.frame(word = names(tv),freq=tv)
wordcloud(words = td$word,freq = td$freq,max.words =600 ,random.color = TRUE,random.order=FALSE,rot.per=0.15,
colors=brewer.pal(8, "Dark2"))